Only Amazon and Tesla made gains in the first November trading session. The split does of course raise the question of how Big Tech is going to turn its massive AI investments into real returns.
The first trading day of November was a mixed bag for the Magnificent Seven. Amazon up 2.20% on Monday the 2nd of November, Tesla up 1.00%. Microsoft down 2.83%, Apple down 1.83%, Meta down 1.34%, Nvidia down 1.26% and Alphabet down 0.30%. The Nasdaq Composite was down 1.1% on that day. That’s just one day’s trading of course, not the whole month, but it does raise the question of why investors are treating the US’s biggest tech companies so differently?
For years the Magnificent Seven have been talked about as a single bet on innovation and growth. Their businesses are in fact very different today as AI turns software bets into massive capital investments. Cloud revenue can go through the roof whilst the cash left over after you’ve built the data centres can go down the pan. Investors are increasingly going to have to assess what each company is actually making on the capital it’s committed rather than just how fast its sales are growing.
Wall Street Is Watching AI Capex. The Real Bill Is Still Coming →
Microsoft generated $18.6 billion in free cash flow in the latest fiscal quarter even though its shares were the worst performer in the group on Monday. Meta recorded $34.91 billion in capital expenditure for the quarter (including finance lease principal payments). Alphabet’s Google Cloud revenue more than doubled to $30.7 billion but Alphabet also reported negative $9.1 billion in free cash flow for the quarter. The numbers are different, but you have to look at profits and cash requirements together.
An AI data centre built today could well be serving paying customers for many years. That makes the substantial current spend perfectly rational. The counterpoint is of course that the equipment will depreciate, power and maintenance will cost money and some long term contracts will go on regardless of whether demand actually meets the forecasts. Technology can transform an industry without every single investment in that technology necessarily delivering an attractive financial return. That’s the tension that investors are now beginning to price in.
Nvidia and Palantir Could Still Be the “Amazon Stocks” of the AI Era →
Amazon is a perfect example of why you can’t just say that Wall Street doesn’t like spending on AI. Its stock was the leader of the seven on Monday. Amazon still expects to spend around $220 billion on capital expenditure this year. It reported a $42.4 billion free cash flow outflow for the trailing twelve months. At the same time AWS sales were up over 39% year on year to around $46 billion in the third quarter. There’s clear revenue growth from its customers even if the infrastructure spend is still extraordinary.
In the October 29 earnings release, CEO Andy Jassy stated, “Our AWS investment in functionality and infrastructure is working.” That’s a management judgement, not necessarily evidence that every project is going to deliver an adequate return. Nvidia, which supplies much of the hardware underpinning the AI expansion, was down 1.26% on Monday. Another reminder that strong demand for compute and investor confidence in the tech valuation are not necessarily two things that go hand in hand.
Tesla’s $1.29 Trillion Valuation Is Getting Strange →
Tesla’s 1.00% gain is a very different investment story. It’s based on the vehicle manufacturing business, the energy storage business and the expectations around autonomous driving and robotics rather than just cloud. A positive day in the market is simply investors putting a higher price on those various prospects on Monday. It in no way implies that the future robotaxi or robotics businesses have in fact become maturely profitable or that their valuations have been validated by cash flow.
The best counter-argument is that the whole group may have been over-punished. Deutsche Bank has become more upbeat on U.S. tech stocks following the recent selloff, suggesting that much of the risk from AI spend was already priced in. That’s a very important point. A stock can fall simply because investors were expecting too much, even if the business underneath is sound and growing.
Investors should now be looking at cloud and AI revenue against capex, depreciation, financing costs and free cash flow over a number of quarters. One day doesn’t define the November winners and Tuesday’s trading can absolutely alter that. The real lesson is that the Magnificent Seven no longer have one common investment thesis. The real question is not which of them can spend the most on AI, but which can actually generate sustainable cash after the bills have been paid.